PROJECT SUMMARY/ABSTRACT The ability to represent recursive structures, or those which allow elements to be embedded within elements of the same kind (e.g. ?[ { } ]?), is thought to be a critical feature of human syntax and has been implicated in the development of a number of human capacities, such as language and mathematics. However, it is currently unknown what processes are used to represent recursive structures, or when the ability to represent recursive structures develops in young children. The overall goals of the proposed studies are to trace the developmental trajectory of the ability to represent recursive structures in children, to rule out deflationary accounts for previous successes, and to determine the type of mechanism used to represent recursive structures. The proposed studies will use a non-linguistic sequencing task to test whether children of different ages (3- to 6-years-old) and adults can represent abstract recursive sequences and generate new, untrained instances of these structures. Subjects will be trained to touch a series of images in a center-embedded, recursive order and then will be tested on their ability to generalize this to untrained lists. Additionally, the proposed studies will control for associative learning mechanisms that previous studies have failed to account for. Lastly, to test the nature of children?s and adults? representations of complex, recursive structures, children?s and adults? ability to learn and generate new instances of different types of structures (center- embedded & cross-serial dependent) will be tested. Different proposed mechanisms for representing recursive structures as well as other structured sequences (stacks and queues) make different predictions about which should be easier and faster to generate. Understanding how children and adults perform on these tasks will shed light on the underlying mechanisms used to generate recursive sequences. These findings will not only provide novel insight into the development of an essential component of human thought, it will provide a foundation for future efforts to understand the underlying cognitive components of language deficits such as those seen in Specific Language Impairment. As the applicant?s previous research has been focused on comparative cognition, the proposed fellowship training plan is targeted at providing a training opportunity to learn developmental psychology methods in an ideal environment, the Harvard Laboratory for Developmental Studies.